Ceramic materials have been utilized in a variety of high temperature and corrosive environments in recent years. Generally, these heat-resistant articles are produced in brick form for the lining of industrial furnaces in the steel, glass, and smelting industries Although castable refractories have been made, they are generally limited to structural applications requiring common geometric configurations such as rounds, squares, and rectangles, due to the ceramic article's tendency to exhibit brittle fracture upon formation or later machining. Although many mechanisms are known to lead to eventual fracture, including the nature of the material selected, the nature of the applied stress, thermal stresses, strain rates, and environmental factors, no viable solution to reducing brittle fracture susceptibility has yet been isolated. A particularly troublesome factor leading to brittle fracture has been the formation of temperature gradients along different surfaces and areas of the ceramic body which frequently lead to thermal shock during processing. This fracture phenomena has heretofore deterred successful use of heat-resistant ceramic articles in gas turbines, ram-jet engines, missiles, nuclear reactors and various other high temperature processes and operations, especially when complex shapes or large body sizes particularly susceptible to temperature gradients are required.
A number of methods have been attempted to produce heat-resistant ceramics of irregular shape and size. These articles are frequently formed from a slurry consisting of fibers, a refractory compound, and a binder, which is molded into the desired shape and then heat treated to form a composite article as taught in Gibson et al. U.S. Pat. No. 3,766,000, and Kajima et al. U.S. Pat. No. 4,158,687. These heat-resistant articles frequently exhibit weak mechanical shock resistance and relatively weak mechanical strength and corrosion resistance at high temperatures. Moreover, stresses due to temperature gradients along the irregularly shaped portions of the ceramic body generally result in thermal shock, which is not dissipated absent the ability to plastically deform, therefore resulting in failure or thermal fatigue. Finally, the very slow step-up heat treating process required to avoid irregular changes in the thermal expansion or contracting of the structure results in a time consuming and expensive process.
In light of these drawbacks there have been developed cermet composite materials, which improve the ductility of brittle ceramic materials by combining them with ductile metals. Although use of metals with a low coefficient of thermal expansion minimizes the thermal shock problem, these same metals easily oxidize and soften at high temperatures, making instances of actual applications in highly corrosive and elevated temperature environments acutely limited.
Recent technological advances have shown that ceramic articles can be altered to absorb energy and deform plastically before fracture through the introduction of a second phase material into the ceramic matrix. This second phase strengthens the ceramic material and increases the fracture toughness by inhibiting crack propagation through the article. Some development work is believed to be underway on composites utilizing long fibers (length/diameter greater than 100). These composite developments are oriented towards very high performance applications where cost is secondary. However, in applications where cost is a significant factor, as in the automotive industry, particle composites or short fiber composites would be preferred so as to avoid time-consuming and expensive fiber preparation techniques.
Notwithstanding this increased attention directed towards fiber and ceramic composite materials, the prior art still lacks a single step processs for producing near-net shape composite articles consisting of a second phase embedded within the base matrix. As disclosed herein in the instant application, however, a composite article may be formed in a single step process from the coupling of a chemical vapor deposition (CVD) produced matrix with a fine particle second phase embedded within the matrix Such articles are formed at high deposition rates and may obviate the above-described prior art disadvantages.
Although the formation of composite articles by entraining solid particles in a CVD reactant gas stream according to the present invention is believed unknown in the art, the direct application of ceramic materials to various substrates by CVD is well known. In Veltri et al., U.S. Pat. No. 4,476,178 the coating of carbon-carbon materials with an outer layer of CVD silicon carbide is disclosed. Depositing a layer of CVD ceramic material on a fugitive carbon form which is later burned away leaving a CVD composed heat exchanger is described in Danis, U.S. Pat. No. 4,488,920 while Galasso et al U.S. Pat. No. 4,373,006 utilizes CVD silicon carbide to thinly coat carbon fibers for insulation purposes. In such conventional processes, as described above, however, virgin CVD material is applied independent of other materials and solely as a thin coat or outer layer to an already self-sufficient substrate. Additionally, the slow deposition rate encountered in prior art CVD processes, such as about 10 mils/hour for silicon carbide deposited from the decomposition of methyltrichlorosilane and hydrogen, seriously impairs the cost-effectiveness of such an application. Moreover, a thin ceramic coating of CVD material is not seen to significantly enhance either the physical characteristics or corrosion resistance of the final product as is accomplished by the present invention.